tmp.py

import re

normal_regex = '[$|£]\s*\S+\s*(?:million)?'

def tmp(b):
    # ['$910,0001955  ', '$350,000 ']
    normal = re.findall(normal_regex, b)
    return normal

def cal_average_budget(budget_list):
    li = []
    for b in budget_list:
        price = 0
        normal = re.findall(normal_regex, b)
        for i in normal:
            cur = i[:1]
            i = i[1:].replace(',', '').replace('，', '')
            for k in unit_dict.keys():
                if k in i:
                    i = i.replace(k, '')
                    for c in split_c_list:
                        if c in i:
                            l = [float(o) for o in i.split(c)]
                            i = sum(l) / len(l)
                            break
                    price = float(i) * unit_dict[k] * multiple_dict[cur]
            if not price:
                price = float(i)
        li.append(price)
    return round(sum(li) / len(li), 2)



# 1954  (27th) - On the Waterfront - $910,000
# 1955  (28th) - Marty - $350,000 [ 1 ] [ 2 ]

lis = [
    # '$910,0001955  '
    '1954  (27th) - On the Waterfront - $910,000',
    '1955  (28th) - Marty - $350,000 [ 1 ] [ 2 ]',
    '1972  (45th) - The Godfather - $6–7 million [ 1 ] [ 2 ]',
]
for l in lis:
    print(cal_average_budget(l))